One Year or One Day?: Adjusting Estimates of Parole and Probation Revocation for Varying Sentence Lengths

• In Event: Applications of Analytics to Community Corrections in Washington, DC

Abstract

How much do changes in a priori length of community correction sentences over time bias estimates of the risk of revocation to incarceration? If comparable offenders who would have served a third year of community supervision in one cohort year finishes supervision successfully in a different year, the overall revocation rate would decline unrelated to any change in offender or community correction agency behavior. This study of recidivism in the District of Columbia uses inverse probability weighting to adjust Kaplan-Meier estimates of the cumulative risk of revocation when the community correction sentence is shorter than the desired follow-period.

Author

• Carroll Ganier, Court Services and Offender Supervision Agency